Design, synthesis and in vivo evaluation of 3-arylcoumarin derivatives of rhenium(I) tricarbonyl complexes as potent antibacterial agents against methicillin-resistant Staphylococcus aureus (MRSA)

Article abstract of DOI:10.1016/j.ejmech.2020.112533

We have prepared a series of ten 3-arylcoumarin molecules, their respective fac-[Re(CO)₃(bpy)L]⁺ and fac-[Re(CO)₃(L?L)Br] complexes and tested all compounds for their antimicrobial efficacy. Whereas the 3-arylcoumarin ligands are virtually inactive against the human-associated pathogens with minimum inhibitory concentrations (MICs) > 150 μM, when coordinated to the fac-[Re(CO)₃]⁺ core, most of the resulting complexes showed remarkable antibacterial potency. Several rhenium complexes exhibit activity in nanomolar concentrations against Gram-positive pathogens such as Staphylococcus aureus strains, including methicillin-resistant S. aureus (MRSA) and Enterococcus faecium. The molecules do not affect bacterial cell membrane potential, but some of the most potent complexes strongly interact with DNA, indicating it as a possible target for their mode of action. In vivo studies in the zebrafish model showed that the complexes with anti-staphylococcal/MRSA activity were non-toxic to the organism even at much higher doses of the corresponding MICs. In the zebrafish-MRSA infection model, the complexes increased the survival rate of infected fish up to 100% and markedly reduced bacterial burden. Moreover, all rescued fish developed normally following the treatments with the metallic compounds.

Full text of DOI:10.1016/j.ejmech.2020.112533

Journal Pre-proof
Design, synthesis and in vivo evaluation of 3-arylcoumarin derivatives of rhenium(I)
tricarbonyl complexes as potent antibacterial agents against methicillin-resistant
Staphylococcus aureus (MRSA)
Sara Nasiri Sovari, Sandra Vojnovic, Sanja Skaro Bogojevic, Aurelien Crochet,
Aleksandar Pavic, Jasmina Nikodinovic-Runic, Fabio Zobi
PII:
S0223-5234(20)30505-5
DOI:
https://doi.org/10.1016/j.ejmech.2020.112533
EJMECH 112533
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 22 April 2020
Revised Date: 29 May 2020
Accepted Date: 31 May 2020
Please cite this article as: S.N. Sovari, S. Vojnovic, S.S. Bogojevic, A. Crochet, A. Pavic, J. Nikodinovic-
Runic, F. Zobi, Design, synthesis and in vivo evaluation of 3-arylcoumarin derivatives of rhenium(I)
tricarbonyl complexes as potent antibacterial agents against methicillin-resistant Staphylococcus aureus
(MRSA), European Journal of Medicinal Chemistry, https://doi.org/10.1016/j.ejmech.2020.112533.
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition
of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of
record. This version will undergo additional copyediting, typesetting and review before it is published
in its final form, but we are providing this version to give early visibility of the article. Please note that,
during the production process, errors may be discovered which could affect the content, and all legal
disclaimers that apply to the journal pertain.
© 2020 Elsevier Masson SAS. All rights reserved.

Design, synthesis and in vivo evaluation of 3-arylcoumarin derivatives
of rhenium(I) tricarbonyl complexes as potent antibacterial agents
against methicillin-resistant Staphylococcus aureus (MRSA)
1
2
2
1
Sara Nasiri Sovari , Sandra Vojnovic , Sanja Skaro Bogojevic , Aurelien Crochet , Aleksandar
Pavic 2*, Jasmina Nikodinovic-Runic 2 and Fabio Zobi 1*
1
Department of Chemistry, University of Fribourg, Chemin du Musée 10, 1700 Fribourg,
Switzerland.
2
Institute of Molecular Genetics and Genetic Engineering, University of Belgrade, Vojvode Stepe
444a, 11042 Belgrade 152, Serbia.
* Correspondence: fabio.zobi@unifr.ch (F.Z.), sasapavic@imgge.bg.ac.rs (A.P.),
1

ABSTRACT: We have prepared a series of ten 3-arylcoumarin molecules, their respective fac-
[Re(CO)3(bpy)L]+ and fac-[Re(CO)3(L L)Br] complexes and tested all compounds for their
antimicrobial efficacy. Whereas the 3-arylcoumarin ligands are virtually inactive against the human-
associated pathogens with minimum inhibitory concentrations (MICs) > 150 µM, when coordinated to
the fac-[Re(CO)3]+ core, most of the resulting complexes showed remarkable antibacterial potency.
Several rhenium complexes exhibit activity in nanomolar concentrations against Gram-positive
pathogens such as Staphylococcus aureus strains, including methicillin-resistant S. aureus (MRSA)
and Enterococcus faecium. The molecules do not affect bacterial cell membrane potential, but some
of the most potent complexes strongly interact with DNA, indicating it as a possible target for their
mode of action. In vivo studies in the zebrafish model showed that the complexes with anti-
staphylococcal/MRSA activity were non-toxic to the organism even at much higher doses of the
corresponding MICs. In the zebrafish-MRSA infection model, the complexes increased the survival
rate of infected fish up to 100 % and markedly reduced bacterial burden. Moreover, all rescued fish
developed normally following the treatments with the metallic compounds.
2

INTRODUCTION
The increase of antimicrobial resistance (AMR) to current clinical drugs and the lack of new
antibiotics to combat emerging resistant strains have become major concerns to human health in this
century [1, 2]. The AMR crisis involves any class of antibiotics including second- and third-line
antibiotics once considered the last resort drugs to tackle common infections, leading to persistence
and spreading of multidrug-resistant (MDR) microbes [3]. Consequently, about 10 million deaths
caused by infectious diseases treatment failure could be reached in 2050 [4]. Researchers and
clinicians are faced with the urgent need to develop novel scaffolds and effective antibacterial drugs
to overcome AMR emergence and spreading of MDR strains [5, 6]. The opportunistic human
pathogens that exhibit multidrug resistance, like Enterococcus faecium, Staphylococcus aureus,
Klebsiella pneumoniae, Acinetobacter baumannii, Pseudomonas aeruginosa, and Enterobacter spp.,
known as ESKAPE pathogens, are considered as a serious global health issue [7, 8]. Among them,
methicillin-resistant Staphylococcus aureus (MRSA) represents the leading cause of nosocomial and
community-acquired infections, being recognized by the World Health Organisation (WHO) as the
antibiotic-resistant pathogen of high priority. The bacterium is the principal causal agent of
complicated skin and soft tissue infections [9], serious bloodstream infections [10], infective
endocarditis, and pneumonia [11, 12]. According to the Centers for Disease Control and Prevention
(CDC), in 2017 almost 120 000 cases of bloodstream infections were caused by S. aureus in the
United States, resulting in almost 20 000 deaths [13]. Current clinical therapy for MRSA infections
relies on the use of vancomycin and daptomycin as the first-line antibiotics, while linezolid represents
the drug of last resort [14]. However, MRSA strains resistant to each of these medicines have already
been reported [15], posing a high demand for novel therapeutic options to combat MRSA-related
infections.
For decades, lead structures in the pharmaceutical industry and most antimicrobial
compounds developed by medicinal chemists were almost exclusively purely organic compounds.
However, studies undertaken in the last 10 years on antimicrobial applications [16-21] of
organometallic and inorganic metal complexes of iridium [22-24], ruthenium [25-29], manganese [30-
33], and gold [21, 34] among others, revealed their strong antimicrobial potency against critical
bacterial and fungal pathogens and distinct mechanisms of action. Some of these molecules
demonstrated activity in the low micromolar or even nanomolar range with no cytotoxicity or
hemolytic activity at much higher concentrations [20, 24]. Interestingly, rhenium complexes, mostly
explored for their anti-tumor properties, were recently shown to possess promising antibacterial
potency [35-41]. In particular, Wenzel, Patra et al. introduced a trimetallic complex comprised of a
ferrocenyl (Fc), a cymantrene and a [(dpa)Re(CO)s] residue (dpa = N,N-bis(pyridine-2-ylmethyl)prop-
2-yn-1-amine) as the main biological active moiety of the construct [36, 37]. MIC values of the
compound against Gram-positive bacterial strains such as B. subtilis and S. aureus strains involving S.
3

aureus ATCC 43300 (MRSA) were in the range of 1.4-21 µM [36]. Siegmund et al. described a
family of antimicrobial N-heterocyclic carbene (NHC) fac-Re(I)(CO)3 complexes containing
unsubstituted benzimidazol-2-ylidene and diimine ligands with MIC values of the complexes between
0.7-2 µM against Gram-positive B. subtilits and S. aureus strains [39]. Recently, Frei et al. reported
the synthesis of three rhenium bisquinoline complexes and their light-induced activity against drug-
resistant S. aureus and E. coli strains, with an increase in antimicrobial activity between 4- to 16-fold
under photo-irradiation than in the dark [40].
While the metal core is of fundamental importance, the nature and coordination mode of the
bound ligands may drastically modulate the overall biological activity (antimicrobial activity and
toxicity) of the given complex and thus its therapeutic application. Coumarins are well-known
antibacterial candidates [42-48], acting via different mechanisms. These include topoisomerase II
DNA (gyrase) [49] or D-alanine-D-alanine ligase (DdlB) [50] inhibition, cell membrane integrity
[51], or cytochrome biosynthesis disruption [52]. Coumarins, naturally occurring and/or synthetic,
also show remarkable pharmacological properties as anticoagulants [53, 54], anti-inflammatory [55],
antioxidant [56], antiplatelet [57], anticarcinogenic [58-63] and antifungal agents [64]. Because metal
coordination of the established drugs can significantly improve their biological activity [32], we
decided to design and synthesize a series of 3-arylcoumarin compounds capable of acting as mono
and bidentate ligands for the fac-[Re(CO)3(bpy)]+ and the fac-[Re(CO)3Br] cores respectively (Fig. 1).
Figure 1. Simplified representation of 3-arylcoumarin mono and bidentate ligands and their
corresponding fac-[Re(CO)3]+ complexes (vide infra for molecular details).
4

The aim of this study was to investigate the antimicrobial potency of the rhenium complexes
and respective ligands against the most prevailing human pathogenic microorganisms and to evaluate
their therapeutic potential for possible applications in anti-infectious therapy. Accordingly, the series
of molecules were tested against Gram-positive and Gram-negative bacterial (E. faecium, S. aureus,
MRSA, L. monocytogenes, P. aeruginosa) and Candida fungal strains (C. albicans, C. glabrata, C.
krusei and C. parapsilosis). The toxicity of the molecules was evaluated in vivo in the zebrafish
(Danio rerio) model and in vitro on human lung fibroblasts (MRC-5 cells). The most promising
complexes, showing a large therapeutic window, were also evaluated for their antimicrobial efficacy
in vivo using the zebrafish-MRSA infection model.
RESULTS AND DISCUSSION
Synthesis and Characterization of Ligands and Complexes.
The ten 3-arylcoumarin-based ligands were prepared according to the synthetic pathways described in
Scheme 1. The synthesis of the 3-arylcoumarin core was achieved by reacting 2-
hydroxybenzaldehyde derivatives with 4-halophenylacetic acid by using potassium acetate in acetic
anhydride. This step involved sequential acetylation of the hydroxyl groups and pyrone ring closure in
a single-pot operation, followed by the hydrolysis of the intermediate using HCl in methanol, which
resulted in the main pharmacophore. The following reactions, such as Buchwald–Hartwig amination,
etherification, copper-catalyzed azide−alkyne cycloadditions (CuAAC) and N-deprotection, were
carried out to obtain the series of mono- and bidentate 3-arylcoumarin ligands L1-L10 in remarkably
high yields (75-95%). Molecules were purified by normal phase column chromatography, fully
characterized by 1H NMR (ESI) and 13C NMR spectroscopy, mass spectrometry, IR spectroscopy, and
1
single-crystal X-ray structure analyses (vide infra). The H NMR spectra of all compounds show a
singlet in the range of 8.0-8.5 ppm associated to the pyrone proton of 4-position and confirmed ring
closure.
The preparation of rhenium complexes ReL1-ReL10 was accomplished via the synthetic protocol
illustrated in Scheme 2. A literature procedure was adapted to our needs [65]. A one-step reaction of
the bidentate ligands (L3, L4, L5, L7 and L9) with rhenium pentacarbonyl bromide resulted in the
desired fac-[Re(CO)3(L#)Br] complexes (where # = 3, 4, 5, 7 and 9). The complexation of 3-
arylcoumarin monodentate ligands (L1, L2, L6, L8 and L10) to fac-[Re(CO)3(bpy)]+ was carried out
stepwise by preparation of the fac-[Re(CO)3(bpy)Br] complex [66], its activation by bromide
-
exchange with triflate (CF3SO3 ) and final displacement of the anion for the desired monodentate
ligands under mild conditions. The cationic fac-[Re(CO)3(bpy)L#]+ species (where # = 1, 2, 6, 8 and
10) were finally isolated as hexafluorophosphate salts by addition of NH4PF6 in methanol. According
to the recent definition of Metzler-Nolte and Cohen [67], these complexes belong to the class F
5

metallofragments based-drugs. All rhenium complexes were also characterized by standard techniques
including single-crystal X-ray structure analysis for ReL2, ReL5 and ReL7 (vide infra). As expected
infrared spectra of rhenium complexes ReL1-ReL10 showed two bands corresponding to the ν(CO)
symmetric A1 and
Scheme 1. Synthesis and chemical structures of 3-arylcoumarin ligands (L1-L10). Conditions: (a) CH3COOK,
Ac2O, at 120 ºC 10 h, under argon/ HCl 2N, MeOH, reflux, 4 h, 90-95%; (b) 4-(bromomethyl)pyridine.HCl,
K2CO3, DMF, NaN3, rt, 4 h, 90%; (c) IM4, CuI , Et3N (anhydrous), 4-bromo-1-butyne, acetonitrile, rt, overnight,
under argon, 85%; (d) 4-picolylamine, Pd2(dba)3, xantphos, Cs2CO3, 1,4-dioxane,100 ºC, overnight for L1, 90%;
(e) N-Boc-ethylenediamine, Pd2(dba)3, xantphos, Cs2CO3, 1,4-dioxane,100 ºC, overnight,/ TFA, DCM, rt, 1.5 h
6

for L3, 80%; (f) 4-picolylamine, 2-picolylamine, Pd2(dba)3, xantphos, Cs2CO3, 1,4-dioxane, 100 ºC, overnight
for L2 and L5, 88% and 85%, respectively; (g) N-Boc-ethylenediamine, Pd2(dba)3, xantphos, Cs2CO3, 1,4-
dioxane, 100 ºC, overnight/ TFA, DCM, rt, 1.5 h for L4, 80%; (h) IM5, K2CO3, acetone, 60 ºC,10 h for L6 and
L10, 75% and 78%, respectively; (i) 1,2-dibromoethane, K2CO3, acetone, 60 C, 8 h, 90%; (j) IM6, NaN3, DMF,
rt, 12 h, 85%; (k) 4-alkynyl-2,2'-bipyridine, CuI, Et3N, acetonitrile, 10 h. for L7, 90%; (l) imidazole, Cs2CO3,
DMF, 110 ºC, overnight for L8, 85%; (m) N-boc-ethylenediamine, DMF(anhydrous), 80 ºC, 4-5h/TFA, DCM,
rt, 1.5 h for L9, 80%.
1
asymmetric E stretching modes at ca. 2030 (s) and 1905 (br) cm-1, respectively (see Table 1). H
NMR spectra in acetonitrile-d3 showed, for fac-[Re(CO)3(bpy)L#]+ complexes, a characteristic low-
field shift of the bipyridine protons, indicative of the successful displacement of the triflate ion. In
coordinating solvents (acetonitrile-d3, methanol-d4, DMSO-d6 or a mixture of water and methanol),
we found no evidence of ligand displacement or solvent exchange for these species over a three days
period.
7

Scheme 2. Synthetic routes of rhenium complexes ReL1-ReL10. Conditions: a) bidentate ligands, toluene, reflux,
12 h, under argon, 90-95%; b) 2,2'-bipyridine, toluene, reflux, 12h, under argon, 98%; c) CF3SO3H, DCM, rt, 1
h, 85%; d) monodentate ligands, MeOH, reflux, 12 h, under argon/NH4PF6, MeOH, rt, 0.5-1 h, 95-98%.
Complexes ReL1-ReL10 show very similar UV-Vis spectra (Table 1). A very pronounced and intense
absorption band centered at 330 nm with a shoulder at 310 nm characterize them (Fig. 2). Because
coumarins are important fluorophores, widely used in both material and biological sciences due to
their inherent physicochemical and photophysical characteristics [69-73], we also investigated the
emissive properties of the complexes. Upon excitation at 360 nm all complexes exhibit emission in
the blue range of the visible spectrum with emission bands between 420–470 nm (Table 1). The only
exception is ReL1, which emits green light (528 nm) under the same conditions.
Table 1. Photophysical properties of ReL1-ReL10.
νCO [cm-1]c
Complexes
Absorption
λmax [nm]a
Emission
λmax [nm]b
528
A1
E
ReL1
ReL2
ReL3
ReL4
ReL5
ReL6
ReL7
ReL8
ReL9
ReL10
330, 310 (sh)
330, 310 (sh)
331, 312 (sh)
335, 307 (sh)
330, 311 (sh)
340, 309 (sh)
338, 328 (sh)
330, 311 (sh)
338, 310 (sh)
342, 311 (sh)
2028
2029
2022
2023
2018
2028
2017
2026
2017
2026
1905
1908
1902
1897
1934
1905
1922
1901
1861
1910
453
470
460
454
428
424
427
425
424
a In CH3CN; sh = shoulder. b In CH3CN; λex = 360 nm. c In KBr.
X-ray crystallography
Single crystals suitable for X-ray diffraction analysis for 3-arylcoumarin ligands L1, L2, L5, L6 and L8
were obtained by slow evaporation of a chloroform or acetonitrile solution of the molecules (Fig. 2).
Similarly, crystals of ReL2 and ReL5 were grown by slow evaporation of acetonitrile:water (2:1) and
dichloromethane solutions, respectively, whilst crystals of ReL7 were obtained by slow diffusion of
pentane into a chloroform solution of the complex (Fig. 3). Ligands L1, L2, L5, L6 and L8 crystallized
in either triclinic (P-1) or monoclinic (P21/c) lattices with 2 or 4 crystallographically independent but
chemically identical molecules in the asymmetric units respectively. With the exception of L2, better
described as L2•CHCl3, no other molecule (i.e. solvents, anion or salts) are present in the asymmetric
units of the other species. Bond distances and angles of the 3-arylcoumarin cores are virtually
identical (within 3σ), indicating that the different chemical derivatizations do not affect significantly
the electronics of the same.
8

Rhenium complexes also crystallized in either triclinic (ReL5, P-1) or monoclinic (P21/c) lattices,
again, with 2 or 4 crystallographically independent but chemically identical molecules in the
asymmetric units respectively. Ligand to ligand π-stacking interactions dominate the crystal packing
of ReL2 and ReL5, but are absent in ReL7 (ESI). Specifically, coumarin ring to bpy π-stacking is
observed in ReL2, interlocking pairs of molecules at a distance of 3.54 Å, while in ReL5 head to tail
fully eclipsed coumarin to coumarin ring π-stacking, pairs the two complexes in the asymmetric unit
at a distance of 3.68 Å. In ReL5, the carbonyl ligand and the bromide ion trans to it are partially
disordered, with relative occupancies of 92 and 8%. The exact same disorder is also present in the
crystal structure of ReL7 where the ligands show partial occupancies of 87 and 13%. No other
-
molecule are present in the asymmetric units of ReL5 and ReL7, while the PF6 counter ion and a
molecule of acetonitrile complete the same of ReL2. In all cases, the rhenium ion is in a
pseudooctahedral environment in which bond lengths and angles fall within the typical range found
for
similar
tricarbonyl
rhenium
complexes
(CCDC
search).
Figure 2. Ortep representation of left: (from top to bottom) ligands L1, L2, L5, and right: (from top to
bottom) L6 and L8. Thermal ellipsoids set at a 30% probability level. Hydrogen atoms, as well as
solvent for L2, are omitted for clarity.
9

Figure 3. Ortep representation of (from top to bottom) ReL2, ReL5 and ReL7. Thermal ellipsoids set at
a 30% probability level. Hydrogen atoms, as well as solvent and counter ion for ReL2 and minor
partitioning for ReL5 and ReL7, are omitted for clarity.
Antimicrobial and Cytotoxic Properties
Antimicrobial activity of ReL1-ReL10 complexes was determined against Gram-positive bacteria, three
Staphylococcus aureus strains including methicillin-resistant S. aureus (MRSA), Enterococcus
faecium and Listeria monocytogenes, Gram-negative bacterial strain Pseudomonas aeruginosa, and
four Candida spp., including C. albicans, C. glabrata, C. krusei and C. parapsilosis (Table 2). In
addition, the therapeutic index as a measure of a selective activity against pathogenic microorganism
in comparison to the host effect was determined by the zebrafish (Danio rerio) model, as a preclinical
animal model for toxicity assessment (Table 2). First, the antimicrobial activity of 3-arylcoumarin
ligands L1-L10 against the same bacteria and fungi was evaluated and minimum inhibitory
concentrations (MICs) were determined. With the exception of L9 (MIC = 20 μM against S. aureus
ATCC 25923 and 10 μM against Candida strains) and L6 (MIC = 13 and 25 μM against C.
parapsilosis and C. glabrata, respectively), none of these ligands exhibited activity, with MICs > 150
μM (ESI).
10

Table 2. Minimum inhibitory concentration (MIC, µg/mL and µM in brackets), therapeutic index (Ti) and
selectivity index (Si) of ReL1-ReL10 complexes against Gram-positive bacterial and fungal strains.
S. aureus
ATCC 43300
S. aureus
NCTC 6571
S. aureus
ATCC 25923
E. faecium
ATCC 6057
Complexes
ReL1
ReL2
ReL3
ReL4
ReL5
0.7 (0.8)/16.7/17.5
0.7 (0.8)/15.7/4.4
7.9 (12.5)/8/1.1
3.9 (6.2)/16/1.2
4.2 (6.2)/16
0.7 (0.8)/16.7/17.5
0.7 (0.8)/15.7/4.4
7.9 (12.5)/8/1.1
7.9 (12.5)/8
2.8 (3.1)/4.2/4.5
2.8 (3.1)/3.9/1.3
> 500
15.8 (25)/4
135.7
2.8 (3.1)/4.2/4.5
1.6 (1.8) /6.9/1.9
> 500
31.5
135.7
4.2 (6.2)/16
ReL6
ReL7
0.8 (0.8)/9.9/26.3
183.7
0.8 (0.8)/9.9/26.3
114.8
3.3 (3.1)/2.5/6.8
> 500
1.9 (1.8)/4.4/11.7
> 500
ReL8
ReL9
0.8 (0.8)/8.5/6
150.2
0.8 (0.8)/8.5/6
187.8
1.5 (1.5)/4.4/3
> 500
6.1
> 500
ReL10
Linezolid
Vancomycin
0.8 (0.8)/24.2/21.9
2 (5.9)/12.5
0.8 (0.8)/24.2/21.9
2 (5.9)/12.5
3.2 (3.1)/6.1/5.6
2 (5.9)/12.5
1(0.7)/nd
0.8 (0.8)/24.3/21.9
nd
nd
1(0.7)/nd
2(1.4)/nd
C. albicans
ATCC 10231
22.5
C. glabrata
ATCC 2001
22.5
C. krusei
ATCC 6258
11.2 / (1.1)
4.5
> 500
> 500
> 500
13.5 / (1.7)
> 500
C. parapsilosis
ATCC 22019
22.5
ReL1
ReL2
ReL3
ReL4
ReL5
ReL6
ReL7
ReL8
ReL9
ReL10
11.2
> 500
> 500
> 500
13.5 / (1.7)
> 500
4.9
> 500
22.5
> 500
> 500
> 500
13.5 / (1.7)
> 500
24.6
> 500
11.2
> 500
> 500
> 500
< 6.7 / (> 3.4)
> 500
4.9
> 500
12.9 / (1.4)
4.9
> 500
12.9 / (1.4)
12.9 / (1.4)
12.9 / (1.4)
Therapeutic Index (Ti= LC50/ MIC) and Selectivity index (Si = IC50/MIC) are given for the most potent
compounds. If is not given the values are ≤ 1. Therapeutic index for vancomycin was not determined due to its
parenteral application and no testing in the zebrafish model. nd – not determined. All complexes are inactive
against Gram-positive L. monocytogenes NCTC 11994 and Gram-negative P. aeruginosa PAO1.
On the other hand, all Re complexes showed good activity against Gram-positive bacteria. Among
these, complexes ReL1-ReL10 exhibited the highest potency against methicillin-resistant S. aureus
ATCC 43300 (MRSA) and S. aureus NCTC 6171, being active in the nanomolar range with MIC
values as low as 0.8 µM (700 - 800 ng/mL). The same species were also active against E. faecium,
with ReL10 exhibiting the highest potency being active at a dose of 0.8 µM (MICs = 800 ng/mL). This
finding is particularly interesting, as S. aureus and E. faecum tend to appear together in hospital
infections and usually can survive in nosocomial environments due to an intrinsic resistance to several
antibiotics [74-76]. In comparison to other rhenium tricarbonyl antimicrobial species tested against
MRSA strains, active ReL# complexes are among the most potent species reported. Their anti-
staphylococcal activity is similar to the N-heterocyclic carbene (NHC) fac-Re(I)(CO)3 compound
containing unsubstituted benzimidazol-2-ylidene and phenanthroline reported by Siegmund et al. [39]
and the rhenium bisquinoline complexes of Frei, if photo-activated at 365 nm (1h irradiation) [40].
In addition, it is important to note that the bipyridyl complexes ReL1, ReL2, ReL6, ReL8 and ReL10
demonstrated 7.6-fold higher antimicrobial activity against MRSA than linezolid (MIC of 2 µg/mL =
5.9 μM) and comparable effects to vancomycin (MIC of 1 µg/mL = 0.7 μM). These are the antibiotics
approved by FDA and recommended by the IDSA guidelines for the treatment of MRSA infections
11

[77]. Vancomycin is the treatment of choice against staphylococcal infections, while linezolid
represents the only alternative to vancomycin for the treatment of MRSA-caused nosocomial
pneumonia in the USA and Europe, proving to be superior over vancomycin in clinical practice [78].
Together with the [(CpXbiph)Ir(TolBig)Br]Br complex [24], ReL10 is the only other example we know
of an organometallic complex showing nanomolar activity against both MRSA and E. faecum.
Against the Gram-negative L. monocytogenes and Gram-negative P. aeruginosa, all Re complexes
were inactive. Similarly, the complexes exhibit moderate (ReL8) or low antifungal activity and low
selectivity towards tested Candida strains (Table 2). Since S. aureus is a leading cause of skin and soft
tissue infections, in order to examine the rhenium complexes as the potential topical agents, we
assessed their cytotoxicity in the model of human lung fibroblasts (MRC-5 cells, Table S2, ESI).
According to the ratio between respective IC50 and MIC values (Table 2), the complexes ReL1, ReL6
and ReL10 showed the best selective index (Si) suggesting their possible application to treat superficial
staphylococcal infection.
S. aureus MRSA membrane potential perturbation and DNA interaction of active complexes.
In order to provide initial clues of the possible targets and mechanism of action of active ReL#
complexes, we measured S. aureus MRSA membrane potential perturbation. The outermost surface of
bacterial cells contains negatively charged structures such as teichoic acid and polysaccharide
elements embedded in the cytoplasmic membrane of Gram-positive bacteria, or lipopolysaccharide in
the outer membrane of Gram-negative bacteria, which are responsible for the initial interaction of the
complexes with bacterial cell surface and subsequent uptake [79]. When tested for their potency to
disrupt bacterial membrane potential, ReL# complexes did not show any significant effect on S. aureus
MRSA cell membrane potential, contrary to carbonyl cyanide m-chlorophenyl hydrazone (CCCP),
used as positive control (Fig. 4). Next, we investigated DNA interaction with ReL# complexes via gel
electrophoresis. Coumarin derivatives are well-known DNA intercalators. Jadoo et al. also reported
that Re(I) species of 4-trifluoromethyl coumarin intercalate in the minor groove of Calf-Thymus-DNA
[80]. Of all tested complexes only ReL1 and ReL2 significantly interacted with double stranded DNA
preventing the interaction of ethidium bromide with lambda bacteriophage DNA (λ-DNA). This
indicates that, at least for ReL1 (Fig. 6) and ReL2 (Fig. S32), bacterial DNA might be a target for their
mode of action. ReL1, in particular, showed the highest affinity for λ-DNA (Fig. 5).
12

Figure 4. Measurement of S. aureus MRSA cell membrane potential after exposure to different
concentrations of selected ReL# complexes (from top to bottom: ½×MIC, 1×MIC and 2×MIC).
Sample fluorescence refers to that of the 3,3'-diethyloxacarbocyanine iodide (DiOC2(3)) dye and
relates to the membrane potential as previously shown.[81] CCCP (carbonyl cyanide 3-
chlorophenylhydrazone) can destroy membrane potential by eliminating the proton gradient and was
used as a positive control. Data relative to ReL1 were normalized to the fluorescence of the complex
in the 1×MIC and 2×MIC experiments.
Figure 5. In vitro interaction of ReL1 with lambda bacteriophage DNA (20 ng/μL, 37 °C, 2 h in 10
mM Tris-Cl, pH 8.5). Concentrations of complex ReL1 are 200, 100, 50, 25, 12.5 and 6.25 µM.
13

In vivo toxicity evaluation
To determine whether ReL# complexes would be safe for human use, we evaluated their toxic
response in the zebrafish model. This is widely used as an animal model for the preclinical toxicity
assessment of bioactive molecules, including a variety of metal complexes (e.g. Ag, Au, Pt, Ru, and
Ir) [82-84]. Owing to their high genetic, physiological and immunological similarity to humans, and
good correlation in response to pharmaceuticals, zebrafish markedly simplifies the path to clinical
trials and reduces failure at later stages of testing [85-87]. To the best of our knowledge, antimicrobial
rhenium(I)-tricarbonyl complexes have not been tested for toxicity in animal studies so far, contrary
to the anticancer species of the same core, thus making their therapeutic profile against microbial
infections largely unknown.
Figure 6. The in vivo toxicity evaluation of complexes ReL1-ReL10 in the zebrafish model. Wild type
(wt) embryos were treated with different doses of tested complexes at 6 hours post fertilization (hpf)
and 72 hpf and evaluated for survival, teratogenicity, cardiotoxicity and the hepatotoxicity at 120 hpf,
respectively (n = 30 per a dose) (A). Normally developed embryos, without signs of hepato- and
cardiotoxicity, after exposure to 10 µM of ReL1 and ReL10 (13 × MICs MRSA for both), 100 µM of
ReL5 (32 × MIC against MRSA) and linezolid at 74 µM (13 × MIC against MRSA) from 72 -120 hpf
are shown (B). The normal (clear) liver in a control embryo is outlined. Tetracycline caused liver
necrosis and darkening (arrow).
The results of our in vivo study showed that ReL# complexes with anti-staphylococcal activity are
non-toxic at much higher doses of the corresponding MICs against MRSA and S. aureus NCTC 6571
(Table 2, Fig. 6A), suggesting their possible application in the treatment of staphylococcal infections.
The most active, bipyridyl complexes ReL1, ReL2, ReL6, ReL8 and ReL10 appeared more toxic than
bidentate complexes ReL3, ReL4, ReL5, ReL7 and ReL9, which correlates with their antimicrobial
activity. The latter set of complexes did not provoke any toxic response in zebrafish embryos even at
a dose as high as 100 µM. The bipyridyl complexes did not cause teratogenic malformations or
cardiovascular side effects (pericardial edema, decreased heart beat rate) in the survived embryos at
any tested dose, while after 5-days exposure induced liver toxicity at doses 27-fold higher of the
14

respective MICs against MRSA (≥ 20 µM). According to the determined LC50 values (Table 2), the
therapeutic indexes (Ti as the LC50 and MIC doses ratio) of antimicrobial ReL# complexes ranged
from 16 to 24, with the highest Ti value for ReL10. Taken together, the data obtained in the zebrafish
model indicate that anti-staphylococcal rhenium tricarbonyl complexes could be applied at much
higher doses than the effective ones considered in vitro, which is of particular significance for
microbial strains that may require higher doses for a therapeutic response.
Antibacterial efficacy in S. aureus-infected zebrafish embryos in vivo system
The antimicrobial efficiency of ReL# complexes was evaluated in vivo using the zebrafish-S. aureus
ATCC 43300 infection model. Zebrafish models of infection are widely employed to investigate the
pathogenesis of various human-associated microorganisms [88, 89] and the differences in virulence
between clinical strains [90, 91], as well as to discover novel antimicrobials and evaluate their
efficacy in vivo [82, 92, 93]. Herein, 30 hpf-old embryos were microinjected with ~1500 bacterial
cells into the bloodstream (the circulation valley) to establish systemic infection. Afterward the
embryos were exposed to different concentrations of the selected Re complexes, and inspected for
their survival over a period of four days post infection (dpi) as a measure of the antimicrobial
efficacy. We chose complexes ReL1, ReL6, ReL8, and ReL10 as the most potent ones, and ReL5 having
the therapeutic index comparable to that of ReL1 and higher than those of ReL6 and ReL8. Their anti-
staphylococcal efficacy was compared to that of the antibiotic linezolid.
In the untreated group, the infected embryos showed the first signs of MRSA infection already 24 h
after bacterial cells injection. Signs included abscess like structures, necrotic yolk/yolk buddy
extension, appearance of pericardial edema, curved body in some cases (Fig. 7B), and embryos
progressively died over time reaching the mortality rate of 60% by 4 dpi (Fig. 7A). The infection
readouts observed herein are very similar to those reported by Prajsnar [94]. On the other side,
treatments with the rhenium complexes successfully rescued infected embryos from lethal
staphylococcal infection, but with different efficiency. The antibacterial efficacy of the rhenium
complexes was evident already 24 h after application, providing markedly higher fish survival
compared to the untreated group (P < 0.01, log-rank test, Fig. 7A) and the absence of necrotic lesions
(abscesses) or pericardial edema (Fig. 7B). Moreover, after 4-days treatments, all rhenium complexes
except of ReL1 appeared more effective against MRSA infection than linezolid at the same dose (P <
0.05, log-rank test). ReL10 appeared to be most potent complex rescuing all infected fish already at ½
× MIC dose, followed by ReL5 which achieved the same protective effect at 1 × MIC and 2 × MIC
doses. Out of four tested complexes, ReL1 demonstrated the lowest efficacy against MRSA in vivo
(active only at ≥ 1 × MIC doses), but again comparable to that of linezolid (P = 0.54 log-rank test),
increasing the survival rate of infected fish up to 90% at the highest applied dose.
15

Figure 7. The tricarbonyl rhenium(I) complexes ReL1, ReL5, ReL6 and ReL10 efficiently rescued the
zebrafish embryos of the lethal MRSA-infection. Wild type (wt) zebrafish embryos were infected
with 1500 cells of S. aureus ATCC 43300 into the circulation valley and incubated at 31 °C (n = 30
per dose). The Kaplan-Meier survival curves of the MRSA-infected embryos upon different doses of
the selected complexes and linezolid are shown (A). Morphology of infected embryos at 1 dpi without
antimicrobial treatment and treated with ReL5 and ReL10 are shown (B). Signs of the MRSA infection
in untreated embryos were visible already at 24 h after bacterial cells injection, such as the abscess-
like structures in yolk buddy extension (arrow), yolk necrosis (asterisk), an appearance of pericardial
edema (arrowhead) and curved body. The antimicrobial efficacy of the applied compounds was
assessed by monitoring fluorescence of bacterial burden in the whole body (C) and determining CFU
per embryos (D) after the 4-days treatment (n = 5-10). Efficacy based on fluorescence data is
expressed as a percentage of inhibition of MRSA proliferation relative to the level for DMSO-treated
control groups. Data are normalized in relation to the control group and shown as the mean ± SD.
Significance in inhibition is indicated with asterisks (*P < 0.05; **P < 0.01; ***P < 0.001).
In order to determine whether the applied complexes also reduced bacterial burden in rescued fish, the
fluorescence intensity of infected embryos correlating with bacterial number and the number of viable
bacteria (CFU) per embryo was assessed after the 4-days treatment. In comparison to the control, we
found that MRSA infection markedly decreased following complexes administration (P < 0.05)
according to the fluorescence decline (Fig. 7C) and the CFU reduction up to 2 log (Fig. 7D).
16

Furthermore, no bacteria were recorded in some embryos receiving 2 × MIC dose of ReL10 (Fig. 7D),
which was significantly more effective than linezolid in MRSA clearance at each of applied dose (P <
0.003).
It is important to note that all rescued fish following ReL# treatment developed without adverse toxic
effects by 4 dpi (120 hpf), including cardiotoxicity, hepatotoxicity and teratogenicity (not shown).
Obtained data regarding anti-MRSA potency in vivo are different from in vitro data. This is not
surprising considering the multicellular complexity of the organism, different conditions and growth
rate of bacterial cells within embryos, the contribution of embryonic innate immune cells (neutrophils
and macrophages) and infected tissue response in the resolving infection, as well as the metabolic
transformation and possible immunomodulatory effect of the absorbed molecules. To the best of our
knowledge, there is still lack of in vivo data demonstrating the antimicrobial activity for this class of
compounds [21]. Our in vivo study in the zebrafish model showed that the rhenium complexes may be
effective antimicrobial agents, with a particular emphasis against MRSA infection. Our complexes
demonstrated better in vivo anti-staphylococcal activity than linezolid, being active in the nanomolar
range and without toxic side effects at higher dose of the effective MIC.
CONCLUSION
In this study we have reported the synthesis, characterization and antimicrobial properties of a series
of 3-arylcoumarin molecules and their respective fac-[Re(CO)3(bpy)L]+ and fac-[Re(CO)3(L L)Br]
complexes. Whereas the organic 3-arylcoumarin ligands are virtually inactive against the pathogens if
administered alone, when coordinated to the fac-[Re(CO)3]+ core, the bactericidal potency of the
resulting complexes is enhanced by 2-3 orders of magnitude. Several of rhenium complexes exhibit
potent in vitro activity against methicillin-resistant Staphylococcus aureus (MRSA)) with minimum
inhibitory concentrations (MICs) in the nanomolar range (700-800 ng/mL). However, in vivo
(zebrafish-S. aureus infection model) complexes proved effective in rescuing MRSA-infected
embryos at lowers doses (as low as 350 ng/mL). Furthermore, the complexes with anti-staphylococcal
/ MRSA activity are non-toxic to the embryos, and are able to increase infected fish rate survival up to
100%. All rescued fish following treatment with the complexes developed without adverse toxic
effects. The molecules do not affect bacterial cell membrane potential, but some of the most potent
complexes strongly interact with DNA, indicating it as a possible target for their mode of action.
Nearly all compounds are ineffective against Candida fungal strains. Overall, our data indicate that
ReL# species could find potential application in antimicrobial therapy against MRSA infections.
Future studies will address their in vivo antimicrobial efficacy against clinical MRSA isolates from
various specimens (e.g. cystic fibrosis, soft tissues, wounds) and decipher their molecular targets and
mode of action.
17

EXPERIMENTAL SECTION
Materials and methods. All chemicals were purchased in reagent or analytical grade from
commercial suppliers (Sigma-Aldrich, Alfa Aesar, TCI, Fluorochem) and used without further
purification. Solvents were either used as received or dried over molecular sieves prior to use. 1H and
13C NMR spectra were recorded on a Bruker Avance III 400 MHz using residual solvent peaks as
internal references. The following abbreviations are used: singlet (s), doublet (d), doublet of doublets
(dd), triplet (t), doublet of triplets (td), quintuplet (quint), sextuplet (sext), and multiplet (m). High
resolution mass spectrometry was performed on a Bruker FTMS 4.7-T Apex II (BRUKER
DALTONICS GmbH, Switzerland). The following abbreviations are used: high-resolution mass
spectrometry (HRMS), electrospray ionization (ESI). HPLC analyses were performed on a Merck-
Hitachi L7000. The analytical separations were conducted on a Macherey–Nagel Nucleodur PolarTec
column (5 μm particle size, 110 Å pore size, 250 × 3 mm). The preparative separations were
conducted on a Macherey–Nagel Nucleodur C18 HTec column (5 μm particle size, 110 Å pore size,
250 × 21 mm). The flow rate was set to 0.5 ml/min for analytical separations and 5 ml/min for the
preparative ones. The eluting bands were detected at 320 nm. Analytical thin-layer chromatography
(TLC) was performed on commercial silica plates (Merck 60-F 254, 0.25 mm thickness); compounds
were visualized by UV light (254 nm and 366 nm). Preparative flash chromatography was performed
with Merck silica gel (Si 60, 63−200 mesh). IR spectra were recorded on a PerkinElmer Spectrum 100
FT-IR spectrometer. The UV-Vis spectra were recorded on a Jasco V-730 and the emission on a
spectrofluorometer FS5 (Edinburgh Instruments Ltd). Single crystal diffraction collections were done
on Stoe IPDS2 diffractometer (MoKα1 (λ = 0.71073 Å)) equipped with a cryostat from Oxford
Cryosystems. The structure were solved with the ShelXT structure solution program [95] using
Intrinsic Phasing and refined with the ShelXL refinement package [96] using Least Squares
minimisation. All the crystal structures have been deposited at the Cambridge Crystallographic Data
Centre. CCDC numbers 1977993-1978000 contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/structures.
DFT and TDDFT Calculations. All computations were performed with the Gaussian 09 programs.
Geometry optimizations as well as frequency calculations were performed in the gas phase. The
hybrid meta-GGA functional wB97XD [97-101], designed to account for dispersion, was used in
combination with the standard SDD basis sets [102]. For the spin state of the complexes (singlet state
in all cases), the default spin formalism was followed in the calculations and default Gaussian 09
values were adopted for the numerical integration grids, self-consistent-field (SCF) and geometry
18

optimization convergence criteria. Geometries were optimized without symmetry restrictions. The
nature of the stationary points was checked by computing vibrational frequencies in order to verify
true minima. No negative frequencies were observed for the reported values. Electronic transition
energies and oscillator strengths were then calculated at their wB97XD -optimized geometries using
TDDFT. For these calculations the 25 lowest energy electronic excitations were calculated for each
compound, and solvent effects were added via a solvent continuum dielectric model using acetonitrile
as the solvent. The calculated molecular orbitals were visualized using GaussView.
Synthesis and characterization of compounds L1-L10.
General procedure for the synthesis of 3-arylcoumarins. 7-bromo-3-phenyl-2H-chromen-2-one
(IM1), 3-(4-bromophenyl)-2H-chromen-2-one (IM2) and 3-(4-bromophenyl)-7-hydroxy-2H-
chromen-2-one (IM3). All glassware used for the preparation of 3-arylcoumarins was prepared to 300
ºC and reactions were performed in anhydrous conditions. To a solution of 4-bromo-phenylacetic acid
(1.0 equiv.) and the corresponding salicylaldehyde (1.0 equiv.) in Ac2O (anhydrous), CH3CO2K (1.8
equiv.) was added at room temperature (rt) under argon, and the mixture was heated to reflux
overnight. After it had cooled to rt, the reaction mixture was neutralized by using 20% aqueous
NaHCO3. Ethyl acetate was added to the aqueous phase and the organic layer was extracted (3×). The
combined organic layers were washed with distilled water and brine, dried with anhydrous Na2SO4
and finally evaporated under reduced pressure. The crude compound was recrystallized from ethanol
giving the acetoxylated intermediate. IM1, IM2, IM3 were prepared by the hydrolysis of their
acetoxylated counterparts with 2 N aqueous HCl in MeOH and the reaction mixture was refluxed for
3-4h. Then the solution was cooled to rt, and kept in an ice-bath for 20min. The solid product was
filtered and washed with cold distilled water (3×), and dried under vacuum. The desired 3-
arylcoumarin intermediates were obtained in 90-95% of yield (no further purification was necessary).
General procedure for the synthesis of L1, L2, L5. To a solution of IM1 or IM2 (1.0 equiv.) in 1,4-
dioxane under argon, Pd2(dba)3 (0.05 equiv), xantphos (0.1 equiv.) and Cs2CO3 (1.5 equiv.) were
added and the mixture stirred at rt for 5 min. Picolylamine (1.8 equiv.) was then added and the
reaction mixture degassed for 15 min. The temperature was increased to 100 ºC and stirred overnight.
After completion of the reaction, the solvent was evaporated, the residue was dissolved in DCM and
the solution was filtered. Then the filtrate was loaded on silica gel for purification by normal phase
column chromatography. The desired product eluted at 50:50 EtOAc:pentane. The organic layer was
concentrated at reduced pressure to yield the product.
3-(4-((2-aminoethyl)amino)phenyl)-2H-chromen-2-one (L1). Yellow solid, yield 90%, ESI-
MS(m/z) calcd. for C21H16N2O2 328.12, found 328.0 [M + H]+. 1H NMR (400 MHz, DMSO-d6) δ 8.51
(d, J = 5.62 Hz, 2H), 8.06 (s, 1H), 7.72 (d, J = 1.47 Hz, 1H), 7.50 - 7.57 (m, 3H), 7.33 (d, J = 0.98 Hz,
19

4H), 6.75 (s, 1H), 6.63 (d, J = 5.01 Hz, 2H), 4.39 (d, J = 6.24 Hz, 2H). 13C NMR (101 MHz,
chloroform-d) δ 161.0, 153.1, 149.9, 148.8, 148.0, 137.3, 130.7, 129.7, 128.1, 127.5, 124.4, 124.3,
122.1, 120.0, 116.3, 112.6, 77.4, 77.2, 77.0, 76.7, 46.8.
3-phenyl-7-((pyridin-4-ylmethyl)amino)-2H-chromen-2-one (L2). Yellow solid, yield 88%, ESI-
MS(m/z) calcd. for C21H16N2O2 328.12, found 328.0 [M + H]+. 1H NMR (400 MHz, chloroform-d) δ
8.46 - 8.55 (m, 2H), 7.55 - 7.65 (m, 3H), 7.31 - 7.36 (m, 2H), 7.25 - 7.30 (m, 1H), 7.18 - 7.24 (m, 3H),
6.47 (dd, J = 2.26, 8.50 Hz, 1H), 6.38 (d, J = 2.20 Hz, 1H), 4.90 (t, J = 5.81 Hz, 1H), 4.39 (d, J = 5.87
Hz, 2H). 13C NMR (101 MHz, chloroform-d) δ 161.3, 155.9, 150.6, 150.0, 147.7, 140.4, 135.4, 129.1,
128.4, 128.3, 128.1, 122.7, 122.0, 111.2, 110.9, 98.3, 77.3, 77.2, 77.0, 76.7, 46.6.
3-phenyl-7-((pyridin-2-ylmethyl)amino)-2H-chromen-2-one (L5). Yellow solid, yield 85%, ESI-
MS(m/z) calcd. for C21H16N2O2 328.12, found 328.0 [M + H]+. 1H NMR (400 MHz, chloroform-d) δ
8.61 - 8.66 (m, 1H), 7.68 - 7.74 (m, 4H), 7.41 - 7.47 (m, 2H), 7.35 - 7.39 (m, 2H), 7.31 - 7.34 (m, 1H),
7.25 - 7.28 (m, 1H), 6.67 (dd, J = 2.26, 8.50 Hz, 1H), 6.56 (d, J = 2.20 Hz, 1H), 5.68 (br. s., 1H), 4.54
(s, 2H). 13C NMR (101 MHz, chloroform-d) δ 171.1, 161.5, 156.5, 156.1, 151.1, 149.2, 140.7, 136.9,
135.6, 128.9, 128.3, 128.3, 127.9, 122.5, 121.9, 121.8, 111.3, 110.7, 97.9, 60.4, 48.3.
General procedure for the synthesis of L3, L4. To the solution of IM1 or IM2 (1.0 equiv.) in 1,4-
dioxane under argon, Pd2(dba)3 (0.05 equiv.), xantphos (0.1 equiv.) and Cs2CO3 (1.5 equiv.) were
added and the mixture stirred at rt for 5 minutes. Boc-N-ethylendiemine (1.8 equiv.) was then added
and the reaction mixture was degassed for 15 min. The temperature was increased to 100 ºC and
stirred overnight. After completion of the reaction, the solvent was evaporated, the residue was
dissolved in DCM and the solution was filtered. Then the filtrate was loaded on silica gel for
purification by normal phase column chromatography. The desired product eluted at 1:6
EtOAc:pentane. The organic layer was concentrated at reduced pressure to yield the N-boc
intermediates. To prepare L3 and L4, (1.0 equiv.) of the N-Boc-protected intermediate was reacted
with TFA in CH2Cl2 and stirred at rt for 1.5h. The residue was dissolved in DCM and carefully
neutralized by saturated 20 % NaHCO3. The organic layer was then washed with water followed by
brine. The organic layer was concentrated. The crude product was purified by recrystallization in
DCM to yield pure L3 and L4.
3-(4-((2-aminoethyl)amino)phenyl)-2H-chromen-2-one (L3). Yellow solid, yield 80%, ESI-
1
MS(m/z) calcd. for C17H16N2O2 280.33, found 281.1 [M + H]+. H NMR (400 MHz, methanol-d4) δ
7.95 (s, 1H), 7.67 (dd, J = 1.53, 7.64 Hz, 1H), 7.50 - 7.62 (m, 3H), 7.31 - 7.38 (m, 2H), 6.69 - 6.76
13
(m, 2H), 3.28 (s, 1H), 3.23 - 3.30 (m, 2H), 2.84 - 2.92 (m, 2H), 2.13 - 2.15 (m, 1H). C NMR (101
MHz, methanol-d4) δ 161.5, 152.8, 149.5, 137.1, 130.3, 129.2, 129.2, 127.8, 127.6, 126.9, 124.3,
122.6, 120.3, 115.5, 111.8, 111.8, 45.3, 40.2.
20

7-((2-aminoethyl)amino)-3-phenyl-2H-chromen-2-one (L4). Yellow solid, yield 85%, ESI-MS(m/z)
1
calcd. for C17H16N2O2 280.33, found 281.1 [M + H]+. H NMR (400 MHz, methanol-d4) δ 7.82 (s,
1H), 7.57 - 7.58 (m, 1H), 7.55 (s, 1H), 7.28 - 7.36 (m, 3H), 7.23 - 7.27 (m, 1H), 6.60 (dd, J = 2.26,
8.62 Hz, 1H), 6.48 (d, J = 2.08 Hz, 1H), 3.42 (t, J = 6.17 Hz, 2H), 3.07 (t, J = 6.17 Hz, 2H), 13C NMR
(101 MHz, methanol-d4) δ 162.1, 156.0, 151.9, 141.5, 135.5, 129.2, 128.0, 128.0, 127.9, 127.9, 127.5,
121.1, 111.3, 110.6, 96.6, 40.2, 38.2.
General
procedure
for
preparation
of
4-((4-(2-bromoethyl)-1H-1,2,3-triazol-1-
yl)methyl)pyridine (IM5). 4-(Bromomethyl)pyridine hydrobromide (1.0 equiv.) was dissolved in
DMF. Potassium carbonate (1.0 equiv.) was added to the mixture and stirred for 15 min. Sodium
azide (1.5 equiv.) was added to the reaction mixture and it stirred at rt for 4h. Afterwards, DCM and
cold water were added to the mixture and two phases were separated (3×). The combined organic
layers were washed with brine and water, dried over Na2SO4 and then filtered. The solvent was
evaporated under reduced pressure giving 4-(azidomethyl)pyridine (85%), as a yellowish oily residue,
which was used for the next step without further purification. To a solution of 4-
(azidomethyl)pyridine (1.0 equiv.) in acetonitrile (dry), CuI (0.1 equiv.) and Et3N (0.4 equiv.,
anhydrous) were added at rt and the solution stirred for 5 min under argon. 4-bromo-1-butyne (1.05
equiv.) was then added and the reaction stirred under argon at rt for 12 h, after which time a mixture
of ice, brine and DCM was added. The two phases were separated, the organic layers were collected,
washed with water twice, combined, washed with brine and finally dried over sodium sulfate and
filtered. The solvent was evaporated under reduced pressure leaving a greenish residue which was
loaded on silica for flash column chromatography. The desired product eluted at 100:1 EtOAc:MeOH.
The pure product (IM5) was obtained as a yellow liquid (90 %).
General procedure for preparation of L6, L10. To a solution of IM3 (1.0 equiv.) in acetone,
potassium carbonate (8.0 equiv) was added and the solution stirred at rt for 10 min. Then, IM5 (1.0
equiv.) was added to the mixture and heated to reflux (60 ºC) for 10h. Then, the mixture was filtered
and the solvent evaporated. The residue was purified using flash column chromatography. The desired
product eluted at 9:1 EtOAc:pentane. L6 and L10 were obtained (75%) and (78%), respectively.
3-(4-bromophenyl)-7-(2-(1-(pyridin-4-ylmethyl)-1H-1,2,3-triazol-4-yl)ethoxy)-2H-chromen-2-
1
one (L6). White solid, yield 75%, ESI-MS(m/z) calcd. for C25H19BrN4O3 503.35, found 503.0. H
NMR (400 MHz, chloroform-d) δ 8.59 - 8.69 (m, 2H), 7.77 (s, 1H), 7.55 - 7.62 (m, 4H), 7.40 - 7.50
(m, 2H), 7.04 - 7.18 (m, 2H), 6.78 - 6.92 (m, 2H), 5.57 (s, 2H), 4.37 (t, J = 6.42 Hz, 2H), 3.30 (d, J =
12.84 Hz, 2H), 13C NMR (101 MHz, chloroform-d) δ 161.8, 160.5, 155.3, 150.6, 144.9, 143.6, 139.9,
133.8, 131.6, 130.0, 129.1, 123.9, 122.8, 122.2, 122.1, 113.4, 112.9, 101.3, 67.3, 52.7, 25.9.
3-(4-chlorophenyl)-7-(2-(1-(pyridin-4-ylmethyl)-1H-1,2,3-triazol-4-yl)ethoxy)-2H-chromen-2-
one (L10). White-yellowish solid, yield 78%, ESI-MS(m/z) calcd. for C25H19ClN4O3 458.90, found
21

1
459.0. H NMR (400 MHz, chloroform-d) δ 8.49 - 8.57 (m, 2H), 7.67 (s, 1H), 7.54 (br. s., 2H), 7.39
(s, 1H), 7.29 - 7.36 (m, 3H), 6.97 - 7.07 (m, 2H), 6.72 - 6.80 (m, 2H), 5.46 (br. s., 2H), 4.27 (t, J =
6.42 Hz, 2H), 3.20 (t, J = 6.42 Hz, 2H). 13C NMR (101 MHz, chloroform-d) δ 161.8, 160.5, 155.3,
150.6, 144.9, 143.6, 139.9, 133.8, 133.5, 130.0, 129.1, 128.6, 123.9, 122.1, 113.4, 112.9, 101.3, 67.3,
52.7, 25.9.
General procedure for preparation of 7-(2-bromoethoxy)-3-(4-bromophenyl)-2H-chromen-2-
one (IM6). To a solution of IM3 (1.0 equiv.) in acetone, potassium carbonate (2.0 equiv.) was added
and the mixture stirred for 5 min under argon. Then 1,2-dibromoethane (7.0 equiv.) was added to the
solution dropwise in 10 min. The mixture was heated to 60 ºC and stirred overnight. After completion
of the reaction, the solution was filtered and washed with acetone (2×). The filtrate was concentrated
under reduced pressure and the residue was recrystallized in acetonitrile to yield the pure product IM6
(90%).
7
-(2-(4-([2,2'-bipyridin]-4-yl)-1H-1,2,3-triazol-1-yl)ethoxy)-3-(4-bromophenyl)-2H-chromen-2-
one (L7). To a solution of IM6 (1.0 equiv.) in DMF under argon, NaN3 (1.5 equiv.) was added and
stirred at rt for 4-5 h. The mixture was added to cold brine, then EtOAc was used to extract the crude
product (3×) from the aqueous phase. The organic layers were collected and concentrated by reduced
pressure. The oily brown residue was purified by a normal phase column chromatography. The
desired product eluted at 60:40 EtOAc:pentane and 7-(2-azidoethoxy)-3-(4-bromophenyl)-2H-
chromen-2-one (IM7) was obtained (95 %) as a white solid. The final step to synthesize L7 was a click
reaction, to a solution of IM7 (1.0 equiv.) in acetonitrile. CuI (0.1 equiv.), Et3N (anhydrous, 0.4
equiv.) and 4-ethynyl-2,2'-bipyridine (1.0 equiv.) were added and then the mixture was stirred for 12h
at rt. After the completion of the reaction, the solvent was evaporated and the crude product was
purified by flash column chromatography. The product eluted at 99:1 EtOAc:MeOH and yielded to L7
(90%). White solid, yield 90%, ESI-MS(m/z) calcd. for C29H20BrN5O3 566.42, found 566.0. 1H NMR
(400 MHz, DMSO-d6) δ 9.06 (s, 1H), 8.86 (d, J = 2.45 Hz, 1H), 8.70 - 8.79 (m, 2H), 8.38 - 8.46 (m,
1H), 8.25 (s, 1H), 7.98 (dt, J = 1.77, 7.73 Hz, 1H), 7.87 - 7.93 (m, 1H), 7.60 - 7.74 (m, 5H), 7.43 -
7.56 (m, 1H), 7.08 - 7.17 (m, 1H), 6.96 - 7.06 (m, 1H), 4.93 (t, J = 5.01 Hz, 2H), 4.66 (d, J = 10.03
13
Hz, 2H). C NMR (101 MHz, DMSO-d6): δ (ppm) 161.48, 160.14, 156.63, 155.45, 155.22, 150.67,
150.59, 149.76, 144.73, 141.46, 139.47, 137.85, 134.53, 131.61, 130.84, 130.38, 124.89, 124.67,
122.74, 122.01, 121.06, 120.26, 116.47, 113.86, 113.62, 101.57, 67.11, 49.74.
7-(2-(1H-imidazol-1-yl)ethoxy)-3-(4-bromophenyl)-2H-chromen-2-one (L8). To a solution of IM6
(1.0 equiv.) in DMF (anhydrous), imidazole (1.5 eqiuv.) and cesium carbonate (2.0 equiv.) were
added. The solution was heated to 110 ºC and stirred overnight. After the completion of the reaction,
the mixture was added to cold brine, then EtOAc was used to extract the crude product (3×) from the
aqueous phase. The organic layers were collected and concentrated by reduced pressure giving an oily
22

brown residue. The crude product was purified by flash column chromatography. The product eluted
at EtOAc and resulted in L8. White solid, yield 90%, ESI-MS(m/z) calcd. for C20H15BrN2O3 411.26,
found 410.9. 1H NMR (400 MHz, acetonitrile-d3) δ 7.99 (s, 1H), 7.61 - 7.67 (m, 4H), 7.56 - 7.60 (m,
2H), 7.16 (t, J = 1.22 Hz, 1H), 6.96 (t, J = 0.98 Hz, 1H), 6.92 - 6.95 (m, 2H), 4.39 - 4.43 (m, 2H), 4.35
- 4.38 (m, 2H). 13C NMR (101 MHz, acetonitrile-d3) δ 161.6, 160.3, 155.4, 140.7, 137.8, 134.7, 131.4,
130.4, 129.7, 128.8, 123.5, 122.0, 119.7, 117.4, 113.8, 113.0, 101.2, 68.1, 49.0, 45.9.
7-(2-((2-aminoethyl)amino)ethoxy)-3-(4-bromophenyl)-2H-chromen-2-one (L9). To a solution of
IM6 (1.0 equiv.) in DMF, N-Boc-ethylenediimine (1.2 equiv.) was added. The mixture was then
heated to 80 ºC and stirred overnight. After the reaction completion, the N-protected intermediate (1.0
equiv.) was dissoleved in DCM (dry) and stirred at rt, then TFA (3.0 equiv.) was added to the mixture
which was stirred at rt for 1.5h. The residue was dissolved in dichloromethane and carefully
neutralized by saturated 20% NaHCO3. The organic layer was washed with brine followed by water
and then concentrated. The crude product was purified by flash chromatography. The product eluted
at 97:3 DCM:NH4OH in MeOH 5% giving L9. White solid, yield 80%, ESI-MS(m/z) calcd. for
C20H15BrN2O3 403.28, found 403.0. 1H NMR (400 MHz, methanol-d4) δ 7.88 (s, 1H), 7.49 - 7.53 (m,
2H), 7.43 - 7.48 (m, 3H), 6.86 (dd, J = 2.38, 8.62 Hz, 1H), 6.81 (d, J = 2.20 Hz, 1H), 5.40 (s, 1H),
4.08 (t, J = 5.26 Hz, 4H), 3.22 (td, J = 1.65, 3.30 Hz, 1H), 3.18 - 3.25 (m, 1H), 3.16 - 3.28 (m, 3H),
2.90 - 2.97 (m, 2H), 2.90 - 2.98 (m, 2H), 2.75 - 2.79 (m, 2H), 2.70 - 2.79 (m, 3H), 2.69 - 2.75 (m, 2H).
13C NMR (101 MHz, acetone-d6) δ 161.5, 159.7, 155.4, 140.5, 134.6, 131.2, 130.4, 129.7,
123.2, 121.8, 113.8, 113.2, 101.1, 66.6, 56.1, 53.2, 42.9.
Preparation of fac-[Re(CO)3(bpy)L#]PF6 Complexes. To a solution of [BrRe(CO)5] (1.0 equiv.) in
hot toluene, bipyridine (1.0 equiv.) was added and refluxed for 7-9h. After it had cooled to the room
temperature, the reaction mixture was filtered and washed with cold toluene (2×), yielding fac-
[Re(CO)3(bpy)Br] as yellow powder. The complex was suspended in DCM under argon and
trifluoromethanesulfonic acid was added and the mixture stirred for 30 min at rt. fac-
[Re(CO)3(bpy)(CF3SO3)], as a green-yellowish solid, was precipitated by slow addition of diethyl
ether. To a solution of fac-[Re(CO)3(bpy)(CF3SO3)] in MeOH, the monodentate ligand was added and
mixture heated to 65 °C. The reaction was refluxed under argon for 10h. To obtain the final products,
NH4PF6 (6.0 equiv) was added to the solution. After precipitation of the final complex, the mixture
was filtered and washed with water to remove the excess NH4PF6.
fac-[Re(CO)3(bpy)L1]PF6 (ReL1). Yellow solid, yield 95%, ESI-MS(m/z) calcd. for C34H24N4O5Re+
1
(M+) 754.80, found 754.90. IR (solid, νCO cm-1): 2028, 1905. H NMR (400 MHz, acetonitrile-d3) δ
9.22 (dd, J = 0.73, 5.50 Hz, 2H), 8.39 - 8.43 (m, 2H), 8.28 (dd, J = 1.47, 15.89 Hz, 2H), 8.14 - 8.18
(m, 2H), 7.85 - 7.89 (m, 1H), 7.78 (d, J = 1.22 Hz, 2H), 7.62 - 7.66 (m, 1H), 7.53 - 7.59 (m, 1H), 7.49
(d, J = 4.77 Hz, 2H), 7.33 (d, J = 1.10 Hz, 2H), 7.28 - 7.30 (m, 2H), 6.53 (d, J = 4.77 Hz, 2H), 5.40 (t,
23

J = 6.24 Hz, 1H), 4.38 (d, J = 6.24 Hz, 2H). 13C NMR (101 MHz, acetonitrile-d3) δ 195.5, 160.7,
155.9, 155.1, 153.9, 153.1, 152.9, 151.7, 148.1, 141.2, 137.4, 130.8, 129.8, 129.7, 128.9, 128.0, 127.8,
124.9, 124.8, 124.8, 124.5, 123.9, 121.4, 120.3, 115.9, 112.2, 45.3.
fac-[Re(CO)3(bpy)L2]PF6 (ReL2). Yellow solid, yield 95%, ESI-MS(m/z) calcd. for C34H24N4O5Re+
1
(M+) 754.80, found 754.90. IR (solid, νCO cm-1): 2029, 1908. H NMR (400 MHz, acetonitrile-d3) δ
9.22 (dd, J = 0.61, 5.50 Hz, 2H), 8.39 (s, 2H), 8.25 - 8.31 (m, 2H), 8.15 - 8.19 (m, 2H), 7.85 - 7.87
(m, 1H), 7.77 - 7.82 (m, 2H), 7.66 - 7.70 (m, 2H), 7.42 - 7.47 (m, 2H), 7.35 - 7.41 (m, 2H), 7.27 (s,
2H), 6.58 (d, J = 2.20 Hz, 1H), 6.20 - 6.22 (m, 1H), 5.88 (s, 1H), 4.42 (d, J = 6.36 Hz, 2H). 13C NMR
(101 MHz, acetonitrile-d3) δ 195.9, 160.9, 156.0, 155.9, 153.9, 151.9, 151.2, 141.2, 140.9, 136.0,
129.4, 128.9, 128.4, 128.3, 127.9, 124.9, 124.8, 121.6, 111.4, 110.6, 97.1, 45.11.
fac-[Re(CO)3(bpy)L6]PF6 (ReL6). Pale yellow solid, yield 95%, ESI-MS(m/z) calcd. for
1
C38H27BrN6O6Re+ (M+) 929.78, found 928.9. IR (solid, νCO cm-1): 2029, 1906. H NMR (400 MHz,
acetonitrile-d3) δ 9.00 (dd, J = 0.79, 5.44 Hz, 2H), 8.45 (br. s., 2H), 8.36 (d, J = 8.19 Hz, 2H), 8.16 -
8.24 (m, 2H), 8.01 (s, 1H), 7.93 (s, 1H), 7.65 - 7.70 (m, 3H), 7.54 - 7.63 (m, 3H), 6.90 - 6.97 (m, 2H),
6.67 (d, J = 4.77 Hz, 2H), 5.14 (s, 2H), 4.36 (t, J = 5.99 Hz, 2H), 3.41 (t, J = 5.93 Hz, 2H). 13C NMR
(101 MHz, acetonitrile-d3) δ 195.9, 161.6, 160.4, 156.6, 155.3, 153.9, 149.9, 148.6, 142.8, 140.8,
140.7, 134.6, 131.4, 130.4, 129.8, 127.8, 125.9, 124.0, 123.5, 122.6, 122.0, 113.8, 113.0, 101.3, 66.1,
53.4, 27.1.
fac-[Re(CO)3(bpy)L8]PF6 (ReL8). Yellow solid, yield 95%, ESI-MS(m/z) calcd. for
C33H23BrN4O6Re+ (M+) 837.68, found 836.9. IR (solid, νCO cm-1): 2026, 1901. 1H NMR (400 MHz,
acetonitrile-d3) δ 9.16 (dd, J = 0.73, 5.50 Hz, 2H), 8.39 (d, J = 8.19 Hz, 2H), 8.24 (dt, J = 1.53, 7.92
Hz, 2H), 8.02 (s, 1H), 7.73 (ddd, J = 1.22, 5.56, 7.64 Hz, 2H), 7.66 (d, J = 1.83 Hz, 4H), 7.59 (d, J =
9.29 Hz, 1H), 7.46 - 7.54 (m, 1H), 7.50 (d, J = 24.58 Hz, 1H), 7.46 - 7.52 (m, 1H), 7.07 (t, J = 1.47
Hz, 1H), 6.71 - 6.76 (m, 2H), 6.57 (t, J = 1.47 Hz, 1H), 4.25 (d, J = 4.89 Hz, 2H), 4.21 (d, J = 4.89
Hz, 2H). 13C NMR (101 MHz, acetonitrile-d3) δ 195.9, 161.3, 160.6, 156.0, 155.4, 154.0, 141.2,
140.9, 140.7, 134.8, 131.7, 130.6, 130.0, 129.1, 128.8, 124.8, 123.9, 122.4, 122.3, 117.6, 114.2, 113.0,
101.4, 67.3, 47.4.
fac-[Re(CO)3(bpy)L10]PF6 (ReL10). Pale yellow solid, yield 95%, ESI-MS(m/z) calcd. for
1
C37H27ClN7O6Re+ (M+) 885.33, found 884.9. IR (solid, νCO cm-1): 2026, 1910, 1892. H NMR (400
MHz, acetonitrile-d3) δ 9.00 (d, J = 5.50 Hz, 2H), 8.51 (br. s., 2H), 8.36 (d, J = 8.19 Hz, 2H), 8.20 (dt,
J = 1.34, 7.95 Hz, 2H), 7.91 - 8.04 (m, 2H), 7.68 - 7.77 (m, 2H), 7.56 - 7.65 (m, 3H), 7.45 - 7.53 (m,
2H), 6.88 - 6.97 (m, 2H), 6.71 (br. s., 2H), 5.15 (s, 2H), 4.36 (t, J = 6.05 Hz, 2H), 3.41 (t, J = 5.99 Hz,
2H). 13C NMR (101 MHz, acetonitrile-d3) δ 195.9, 161.6, 160.4, 156.6, 155.3, 153.9, 149.9, 148.6,
24

142.8, 140.8, 140.7, 134.6, 131.4, 130.4, 129.8, 127.8, 125.9, 124.0, 123.5, 122.6, 122.0, 113.8, 113.0,
101.3, 66.1, 53.4, 27.1.
Preparation of fac-[Re(CO)3(L#)Br] Complexes. Bromopentacarbonylrhenium(I) (1.0 equiv.) was
dissolved in hot toluene (110 °C) under argon. The diimine ligand (1.0 equiv.) was added, and the
solution was refluxed overnight. The mixture was cooled to room temperature, and if a precipitate had
formed, it was isolated by centrifugation, washed with cold toluene and cold diethyl ether, and dried
in vacuo. Otherwise, the solvent was evaporated, and the crude compound was chromatographed by
flash chromatography on silica gel using ethyl DCM/MeOH (100:3, v/v) as eluent to give the pure
products described below.
fac-[Re(CO)3(L3)Br] (ReL3). Light brown solid, yield 95%, ESI-MS(m/z) calcd. for
C20H16BrN2O5Re (M-Br)+ 550.56, found 550.8. IR (solid, νCO cm-1): 2023, 1902, 1876. 1H NMR (400
MHz, acetone-d6) δ 8.08 (s, 1H), 7.73 - 7.78 (m, 2H), 7.64 - 7.68 (m, 1H), 7.47 - 7.52 (m, 1H), 7.24 -
7.28 (m, 4H), 5.86 - 5.87 (m, 1H), 5.07 (d, J = 5.14 Hz, 1H), 3.94 (d, J = 7.70 Hz, 1H), 3.37 - 3.48 (m,
3H), 3.13 - 3.26 (m, 1H). 13C NMR (101 MHz, acetone-d6) δ 159.7, 153.6, 149.5, 139.9, 132.1, 131.5,
129.4, 128.5, 126.7, 124.5, 119.9, 119.1, 115.9, 53.8, 42.9.
fac-[Re(CO)3(L4)Br] (ReL4). Dark cream solid, yield 93%, ESI-MS(m/z) calcd. for C20H16BrN2O5Re
1
(M-Br)+ 550.56, found 550.8. IR (solid, νCO cm-1): 2023, 1897. H NMR (400 MHz, acetone-d6) δ
8.16 (s, 2H), 7.82 - 7.85 (m, 2H), 7.80 - 7.82 (m, 2H), 7.78 - 7.80 (m, 2H), 7.50 (t, J = 1.77 Hz, 1H),
7.47 - 7.48 (m, 2H), 7.46 - 7.47 (m, 2H), 7.44 - 7.45 (m, 1H), 7.42 - 7.44 (m, 1H), 7.31 - 7.34 (m, 4H),
6.21 - 6.28 (m, 2H), 5.27 (d, J = 5.26 Hz, 2H), 4.07 - 4.15 (m, 2H), 3.62 - 3.67 (m, 4H), 3.59 (dt, J =
3.06, 6.05 Hz, 2H), 3.32 - 3.43 (m, 2H). 13C NMR (101 MHz, acetone-d6) δ 159.5, 154.1, 152.2,
139.6, 135.2, 129.2, 128.6, 128.5, 128.2, 126.9, 117.1, 116.6, 105.9, 53.3, 43.0.
fac-[Re(CO)3(L5)Br] (ReL5). Brown-reddish solid, yield 90%, ESI-MS(m/z) calcd. for
C24H15BrN2O5Re (M-Br)+ 597.60, found 598.9. IR (solid, νCO cm-1): 2018, 1934. 1H NMR (400 MHz,
acetone-d6) δ 8.77 (d, J = 5.26 Hz, 1H), 8.01 - 8.11 (m, 2H), 7.78 - 7.84 (m, 1H), 7.70 - 7.76 (m, 1H),
7.65 (d, J = 1.22 Hz, 2H), 7.49 (br. s., 1H), 7.33 - 7.34 (m, 1H), 7.29 - 7.32 (m, 2H), 7.25 - 7.29 (m,
2H), 7.03 (br. s., 1H), 5.10 - 5.27 (m, 2H). 13C NMR (101 MHz, acetone-d6) δ 162.1, 159.9,
153.0, 151.9, 140.0, 139.5, 135.1, 131.7, 129.7, 129.0, 128.6, 128.2, 127.1, 125.5, 122.9,
116.3, 106.0, 103.9, 59.0.
fac-[Re(CO)3(L7)Br] (ReL7). Yellow solid, yield 95%, ESI-MS(m/z) calcd. for C32H20Br2N5O6Re
(M-Br)+ 836.65, found 835.8. IR (solid, νCO cm-1): 2018, 1922, 1889. 1H NMR (400 MHz, DMSO-
d6) δ 9.20 (s, 1H), 9.11 (d, J = 1.10 Hz, 1H), 9.06 (d, J = 5.87 Hz, 2H), 8.91 (d, J = 8.19 Hz, 1H), 8.39
(dt, J = 1.41, 7.92 Hz, 1H), 8.25 (s, 1H), 8.20 (dd, J = 1.59, 5.87 Hz, 1H), 7.75 - 7.83 (m, 1H), 7.61 -
7.73 (m, 5H), 7.14 (d, J = 2.20 Hz, 1H), 7.02 (dd, J = 2.45, 8.68 Hz, 1H), 5.01 (t, J = 4.65 Hz, 2H),
25

4.65 (t, J = 4.77 Hz, 2H). 13C NMR (101 MHz, DMSO-d6) δ 161.4, 160.1, 156.5, 155.6, 155.2, 154.2,
153.6, 143.2, 141.8, 141.4, 140.7, 134.5, 131.6, 130.8, 130.4, 128.5, 126.6, 125.0, 123.0, 122.8, 122.0,
120.2, 113.9, 113.6, 101.6, 67.3, 50.0.
fac-[Re(CO)3(L9)Br] (ReL9). Cream-yellowish solid, yield 90%, ESI-MS(m/z) calcd. for
1
C22H17Br2N2O6Re (M) 751.40, found 749.90. IR (solid, νCO cm-1): 2017, 1898, 1862. H NMR (400
MHz, acetone-d6) δ 8.15 (s, 1H), 7.75 (s, 1H), 7.70 (d, J = 8.44 Hz, 1H), 7.62 (s, 2H), 7.01 - 7.07 (m,
2H), 5.05 (d, J = 3.42 Hz, 1H), 4.57 - 4.66 (m, 1H), 4.47 - 4.56 (m, 1H), 4.13 (br. s., 1H), 3.84 - 3.92
(m, 1H), 3.73 - 3.83 (m, 1H), 3.52 - 3.68 (m, 2H), 3.17 - 3.32 (m, 2H), 2.78 - 2.89 (m, 2H). 13C NMR
(101 MHz, methanol-d4) δ 162.3, 160.8, 155.1, 140.8, 134.1, 131.0, 131.0, 129.9, 129.9, 129.4, 122.7,
121.9, 113.4, 113.0, 100.5, 67.7, 53.4, 49.7, 39.9.
Antimicrobial Assays. The minimum inhibitory concentrations (MICs) against a panel of bacteria
and Candida spp. (all obtained from the National Collection of Type Cultures (NCTC) and the
American Type Culture Collection (ATCC)) were determined according to the standard broth
microdilution assays, recommended by CLSI (Clinical and Laboratory Standards Institute. Methods
for Dilution Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically; Approved
Standard - Tenth Edition M07-A10. CLSI) and EUCAST (European Committee on Antimicrobial
Susceptibility Testing, EUCAST antifungal MIC method for yeasts, v 7.3.1), respectively. The test
strains studied were cultured in Luria-Bertani broth (Biolife Italiana S.r.l., Milano, Italy) and RPMI
1640 medium (Gibco) and diluted to give the concentration of 5 × 105 CFU/mL for bacterial and 1 ×
105 CFU/mL for fungal cells, respectively. The complexes in liquid broth were serially diluted in the
sterile 96-well plate to give the volume of 100 μL. The media solutions with microorganisms were
then dispensed to each well to make the final volume of 200 μL, and the final concentration of
complexes was in the range 0.125−800 μM. All the plates were covered and incubated at 37 °C for 18
h without shaking. Inhibition of bacterial growth was determined measuring absorbance at 600 nm
(OD600), using a Tecan Infinite 200 Pro multiplate reader (Tecan Group Ltd., Männedorf,
Switzerland). The negative control (media only) and positive control (only microorganisms in liquid
broth) on the same plate were used as references to determine the growth inhibition of bacteria.
Samples with inhibition values above 90% were classified as active agents.
In vitro cytotoxicity. In vitro cytotoxicity was determined as antiproliferative activity by (3-(4,5-
dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay [103] on human lung fibroblasts
(MRC-5) obtained from American Type Culture Collection (ATCC) collection. The cells, cultured in
the complete RPMI 1640 medium (Merck, Munich, Germany) as a monolayer (1.104 cells per well),
were incubated with test compounds at concentrations ranging from 3.1 to 100 µM, in humidified
atmosphere of 95% air and 5% CO2 at 37 °C and the cell viability was measured after 48 h. The
extent of MTT reduction was measured spectrophotometrically at 540 nm using Tecan Infinite 200
26

Pro multiplate reader (Tecan Group Ltd., Männedorf, Switzerland) and the cell survival was
expressed as percentage of the control (DMSO treated cells) arbitrarily set to 100%. Cytotoxicity is
expressed as the concentration of the compound inhibiting growth by 50% (IC50).
S. aureus membrane perturbation assessment. The depolarization of plasma membrane potential of
S. aureus ATCC43300 MRSA was determined as described previously by using the membrane
potential-sensitive fluorescent dye DiOC2(3) (3,3'-Diethyloxacarbocyanine Iodide) [81]. Briefly, S.
aureus MRSA was grown to the early exponential phase (OD625nm of 0.3) in LB, washed with PBS
(pH 7.4) and resuspended in PBS containing 1% (w/v) glucose (108 colony-forming units of bacteria
per mL). Bacteria were incubated at 37°C for 15 min, followed by incubation with 3 µM DiOC2(3)
under the same conditions (37°C for 15 min). After the incubation period bacteria were transferred to
96 well plate (100 µl per well) and treated with different compounds. Carbonyl cyanide m-
chlorophenyl hydrazone (CCCP, 0.25 µM) was used as positive control (at final concentration of
0.025 µM). The fluorescence change (excitation 485 nm, emission 630 nm) was measured
immediately after the addition of compounds on a Tecan Infinite 200 Pro multiplate reader (Tecan
Group Ltd., Mannedorf, Switzerland).
DNA Interaction via gel electrophoresis. DNA interaction assay using gel electrophoresis was
conducted according to the previously published procedure [104] using commercial lambda
bacteriophage DNA (300 ng, Thermo Scientific™). DNA solution of the final concentration 20 ng/μL
was incubated with 400 µM, 200 µM, 100 µM of ReL# compounds (in addition, lower concentrations
of ReL1 of 50 µM, 25 µM, 12.5 µM, 6.25 µM were used) in 15 μL reaction volume. After 1 h
incubation at 37 °C, samples were mixed with loading dye and analysed via gel electrophoresis (300
ng DNA per lane, 0.8% agarose gel) with ethidium bromide (EthBr) against a HyperLadder™ 1 kb
DNA Ladder plus (FastGene) at 60 V for 1 h. Gels were visualized and analysed using the Gel Doc
EZ system (Bio-Rad, Life Sciences, Hercules, USA), equipped with the Image Lab™ Software.
In vivo toxicity. Toxicity evaluation of Re(I)-tricarbonyl complexes was carried in the zebrafish
(Danio rerio) model according to the general rules of the OECD Guidelines for the Testing of
Chemicals (OECD, 2013, Test No. 236) [105]. All experiments involving zebrafish were performed in
compliance with the European directive 2010/63/EU and the ethical guidelines of the Guide for Care
and Use of Laboratory Animals of the Institute of Molecular Genetics and Genetic Engineering,
University of Belgrade. Wild type (AB) zebrafish were kindly provided by dr Ana Cvejić (Wellcome
Trust Sanger Institute, Cambridge, UK), raised to adult stage in a temperature- and light-controlled
zebrafish facility at 28 °C and standard 14:10-hour light-dark photoperiod, and regularly fed with
commercially dry flake food (TetraMinTM flakes; Tetra Melle, Germany) twice a day and Artemia
nauplii once daily. Embryos were produced by pair-wise mating, collected and distributed into 24-
27

well plates containing 10 embryos per well and 1 mL embryos water (0.2 g/L of Instant Ocean® Salt
in distilled water) and raised at 28 °C. For assessing lethality, developmental toxicity and
cardiotoxicity, the embryos staged at 6 hours post fertilization (hpf) were exposed to different
concentrations of the tested compounds, and inspected for apical endpoints (Table S1) every day by
120 hpf upon an inverted microscope (CKX41; Olympus, Tokyo, Japan). Linezolid was used as a
reference anti-staphylococcal antibiotic. DMSO (0.25%) was used as negative control. Dead embryos
were counted and discarded every 24 h. Experiments were performed two times using 30 embryos per
concentration. At 120 hpf, embryos were inspected for heartbeat rate, anesthetized by addition of
0.1% (w/v) tricaine solution (Sigma-Aldrich, St. Louis, MO), photographed and killed by freezing at -
20 °C for ≥ 24 h. In addition to developmental and cardio- toxicity, Re(I)tricarbonyl complexes were
evaluated for hepatotoxicity. Wild types embryos were exposed to non-lethal doses of the rhenium
complexes in a period from 72 hpf (a stage when liver is perfused with blood and a metabolically
active) to 120 hpf, and inspected for the hepatotoxicity endpoints (the liver necrosis and darkening,
the yolk necrosis and reduced consumption) as previously described by He et al. [106] . Tetracycline-
hydrochloride was used as a hepatotoxic antibiotic (positive control) [106]. The experiment was
performed two times using 20 embryos per a concentration.
Testing of anti-MRSA efficacy in vivo. To address antimicrobial efficacy of rhenium complexes,
wild type zebrafish embryos were challenged to S. aureus ATCC 43300 infections, according to the
protocol previously described by Prajsnar [89]. Bacteria used to infect embryos were grown overnight
in Brain-heart infusion (BHI) broth, diluted at 1 :100 ration and incubated at 37 °C to reach a mid of
exponential phase. After centrifugation (4500 × g, 10 min), bacterial pellet was washed three times in
PBS and adjusted to OD600 (optical density at 600 nm) of 3×108 cells/mL. Bacterial cells were than
labelled with 2 µM CellTrackerTM RedCMTPX (Thermofisher Scientific) according to the
manufacturer’s instructions. Prior infection, embryos were manually dechorionated at 24 hpf stage
and keep at 28 °C. At 30 hpf, anesthetized embryos were microinjected with 5 nL containing ~1500
labelled MRSA cells into the circulation valley by a pneumatic picopump (PV820, World Precision
Instruments, USA) aimed to establish fast systemic infection. To confirm bacterial numbers in each
injection, a few infected embryos was crashed immediately and the viable counts was determined on
solid BHI agar plates after incubation for 4 days at 37 °C. After injection, embryos were incubated to
recover for two hour at 31 °C, dead embryos were removed, and alive embryos were transferred into
24-well plates containing 1 mL of embryo water and 10 embryos per well. The infected embryos were
treated with three doses (½ × MIC, 1 × MIC and 2× MIC) of the selected complex, and maintained at
31 °C by 120 hpf. Linezolid was used as a positive control. Twenty embryos were used per
concentration. Embryos injected with PBS and treated with 0.25% DMSO were used as a control
group. The survival and development of infected embryos was recorded every day until 120 hpf
(corresponding to fourth days post infection). To evaluate antimicrobial efficacy, bacterial infection
28